Russia: Restoring Mountain Birch Seedlings in a Heavily Polluted Industrial Barren on the Kola Peninsula


Denuded landscapes adjacent to big polluters represent an extremely harsh environment for plants due to a unique combination of soil toxicity and physical stress. A 5-year experiment conducted at two barren sites near the Severonikel nickel-copper smelter at Monchegorsk, northwestern Russia tested whether the survival and performace of seedlings of Mountain birch (Betula pubescens ssp. czerepanovii) can be enhanced by physical sheltering and how large the supposed benefits to survival and performance are in relation to liming. Seedling performance was assessed by measuring growth parameters and chlorophyll fluorescence. Physical sheltering was found to be beneficial in some conditions: when soil characteristics were not so harsh as to cause 100% mortality, sheltering increased both survival and performance of birch seedlings. Although the benefits of liming on seedling performance and survival were stronger than the benefits of sheltering, sheltering may still have its uses in restoration when large-scale liming is not applicable, for example, when the ecological side effects of liming are to be avoided.

Quick Facts

Project Location:
Monchegorsk, Murmansk Oblast, Russia, 67.943134, 32.89195619999998

Geographic Region:

Country or Territory:



Area being restored:
(experimental sites)

Organization Type:
University / Academic Institution


Project Stage:

Start Date:

End Date:

Primary Causes of Degradation

Mining & Resource Extraction, Urbanization, Transportation & Industry

Degradation Description

The Severonikel smelter is one of the top polluters in Europe, predominantly in sulfur dioxide (SO2) and heavy metals (Pearce 1994). In the early 1990s, it emitted annually 200 kt of SO2 and 2.7 and 3 kt of nickel and copper, respectively (Barcan 2002). By 1963, aerial pollution had destroyed forests up to 7 km from the smelter (Pozniakov 1993), and at present, areas of dead forest reach up to 15 km south and 10 km north of the smelter (Kryuchkov 1993; Mikkola 1996b). More than 10,000 km2 have been affected by large-scale environmental deterioration, and vast areas of previously healthy forest within 400 km2 of the smelter have been transformed into secondary open habitats (conventionally named industrial barrens) with extremely low (<5%) vegetation cover and high physical stress (Kryuchkov 1993; Mikkola 1996a; Rigina & Kozlov 1999).

These barrens exhibit a unique combination of environmental characteristics, including: low vegetation cover due to disturbance-related fragmentation of ground-layer vegetation, increased light availability due to forest decline, low mineral nutrition due to displacement of base cations by heavy metals, and harsh microclimate, in particular increased wind speed and deep soil freezing due to thin snow cover (Koroleva 1993; Kryuchkov 1993; Kozlov & Haukioja 1997; Derome & Nieminen 1998; Lukina & Nikonov 1999; Rigina & Kozlov 1999; Winterhalder 2000; Kozlov 2001, 2002; Zvereva & Kozlov 2001). Low soil temperatures (circa -10ºC) during winter have been found to negatively affect Mountain birch growth (Weih & Karlsson 2002), and wind stress has often been considered as a factor decreasing plant performance, also in conjunction with pollution (Carlsson & Callaghan 1991; Hoad et al. 1998; Zvereva & Kozlov 2004).

Mountain birch (Betula pubescens ssp. czerepanovii (Orlova) HÁ¤met-Ahti) is one of the few woody plants surviving in industrial barrens, where it grows as low-stature, multi-stem bushes (Kryuchkov 1993). As the tree line-forming species in northern Europe, Mountain birch is resilient and can withstand the harsh climatic conditions of cold, windy habitats. Even severe pollution, associated with both an increase in the impact of climate stress and a decrease in environmental capacity, was not found to decrease the sexual reproduction of Mountain birch in terms of proportion of reproducing trees, reproductive effort, catkin weight, and seed germinability (Kozlov & Zvereva 2004). However, no natural recruitment of this species has been observed in industrial barrens during the past decades (Kozlov & Haukioja 1999; Rigina & Kozlov 1999). The persistence of Mountain birch populations near strong sources of pollution may, thus, be transient, and a complete decline of birch populations on heavily contaminated soils may be prevented only by artificial reforestation. Past attempts at sowing Mountain birch seeds, however, resulted in 100% mortality by the end of the first growth season (Kozlov 2005a), and the survival of unprotected Mountain birch seedlings planted near the Severonikel smelter is very low, even down to 0 – 10% (Pankratova 1991; Kozlov & Haukioja 1999). Thus, although Mountain birch is a potential species for reforestation of industrial barrens, there is a clear need to find ways to mitigate the impact of diverse stressors co-occurring in industrial barrens in order to assure restoration success.

Reference Ecosystem Description

Due to the influence of the warm Gulf Stream current, the Kola Peninsula has a relatively mild and stable climate and is free of permafrost. The climate is normally cool, with low summer and winter air temperatures: the average temperature in January is -8°C along the northern coast and -12 to -15°C in the centre of the peninsula. Winter is characterized by frequent blizzards, causing large snowdrifts, while summer is short, lasting only a few months, and is generally cool and rainy, with average June temperatures ranging from 8 to 14°C. Three-quarters of all precipitation occurs between June and October, with an annual total of 400 mm. February is the coldest and windiest month, and July is the warmest and most calm. In winter, the sun remains below the horizon between 1 December and 13 January (at Murmansk), and in summer, the sun does not set between 23 May and 21 July (at Murmansk). In the south this period is shorter, only lasting from 2 June to 11 July.

Up to the mid-1930s, descriptions of the Kola Peninsula spoke of virgin pine forest and impenetrable spruce forest with “˜beards’ of pollution-sensitive epiphytic lichens covering the area a few kilometers to the south of the present-day city of Monchegorsk. Indeed, forests cover about half of the peninsula, ranging from northwest to southeast, and are composed of coniferous trees such as Scots pine and Norway spruce, and deciduous trees such as birch, mountain ash, and alder. Prior to construction of the smelter in 1938, mountain birch comprised part of a mixed forest that covered the study area, together with Scots pine and Norway spruce (Bobrova & Kachurin 1936).

Project Goals

To evaluate the effect of physical sheltering on the survival and performance of Mountain birch seedlings planted on severely contaminated industrial barren sites.


The project does not have a monitoring plan.

Description of Project Activities:
Five blocks, circa 8 x 15 m, were established at each site, with nine treatments in each block. From four to seven (median = 5) seedlings were assigned into each treatment in each block at an average distance of 10 cm from one another. The average amount of seedlings was 59 per treatment for a total of 530 experimental seedlings, 243 in site 1 and 287 in site 2. The seedlings were planted at site 1 on 5 June 1999 and at site 2 on 2 June 1999. As a source population, seedlings from a relatively unpolluted site 63 km southeast of the Severonikel smelter were used (lat 67º32'16"N, long 33º57'52"E). The seedlings were 3 - 10 cm tall and were naturally recruited in a belt of sandy soil exposed some 15 years ago by road construction. Care was taken to distribute seedlings of different height evenly among treatments and blocks (i.e., experimental units). The seedlings were planted bare-rooted, with a minimum amount of native soil inserted. In addition to control (C), each block included eight treatments with natural or artificial stress ameliorators: liming (L), boxing (B), a combination of liming and boxing (LB), watering (W), a combination of watering and boxing (WB), partial shelter (PS), full shelter (FS), and stone (St). Liming treatment consisted of a one-time addition of 40 g of granulated (grain size 1 - 3 mm) dolomitic limestone (CaMg(CO3)2, with 17% Ca and 5% Mg; Kemira Growhow Oyj, Helsinki, Finland) into an area of 1 m2 around the seedlings at the time of planting. This was done to achieve a more natural pH in the soil and also to minimize heavy metal uptake by the seedlings. Watering treatment was done at the time of planting and every second day for the first 5 weeks of the experiment, excluding rainy days. Local pond water was used to irrigate the seedlings, with 2 L added to an area of 1 m2 around them. Irrigation was added to check for the effect of drought because in industrial barrens, where shading is very low, soil moisture is reduced on average to 75% relative to undisturbed habitats (Kozlov, unpublished data). Boxing treatment consisted of circling the seedlings with a wooden fence 25 cm high and 25 x 40 cm in surface area. Partial and full shelters were 25-cm-high and 40-cm-wide wooden windbreaks placed between the seedlings and the smelter at a distance of circa 15 cm from the seedlings. Full shelters were solid, whereas partial shelters had 6-cm-wide horizontal gaps in the middle and at the bottom (i.e., 50% permeable). In the stone treatment, seedlings were planted within a 30-cm distance of a natural rock 15 - 90 cm high and 30 - 110 cm wide so that the rock was located between the seedlings and the smelter. Treatments involving boxing, shelters, and stone were designed to improve the microclimate around the seedlings and to see whether a natural (and thus no-cost) shelter would benefit the seedlings, or whether an artificial (and thus costly) shelter is necessary. The effects of the treatments on microclimate were tested in the summer of 1999. Wind velocity in each experimental unit was measured six times between 15 June and 14 August using Kestrel Pocket Wind Meter (Nielsen-Kellerman, Chester, PA, U.S.A.; speed range 0.3 - 40 m/second, accuracy 0.1 m/second or 3% of reading for velocities exceeding 3.5 m/second). Measurements within the study site were performed in a random order, simultaneously by two wind meters placed at a height of 2 m (wvc) and at the level of seedling crowns (wvt), with impellers perpendicular to the wind direction. Average wind speeds were recorded from 10 seconds measurements. The relative wind velocity was calculated as wvt/wvc x 100%. To check whether sheltering had any effect on the amount of rainfall reaching soil near the seedlings, transparent plastic vials (38 and 70 mm in diameter and height, respectively) were placed next to the experimental seedlings, and the amount of water in each vial was checked after episodes of heavy rain, five times in site 1 and seven times in site 2. Illumination was measured six times during the summer with a Panlux electronic luxmeter (Gossen, NÁ¼rnberg, Germany). The measurements were conducted next to seedling crowns (it) and at an open site next to them at a height of 1.5 m (ic) in heavily cloudy days to avoid the effects of random shadows. Relative illumination (it/ic x 100%) was calculated for each pair of measurements. The survivors in each experimental unit were counted five times during the summer of 1999, at the beginning and at the end of growing seasons from 2000 to 2002, and at the end of the growing season in 2004. Performance indexes--i.e. lengths of two largest leaves (without the petiole) and all long shoots (e.g., shoots that grow tens of millimeters long and bear leaves that expand one after another following shoot elongation)--were measured late in the growing season (when leaves were fully matured and shoot growth had ceased) in years 1999 - 2002 and 2004, and seedling heights in 2000, 2001, and 2004. In 2000 and 2001, chlorophyll fluorescence was measured from two intact leaves per seedling in field conditions using a portable plant stress meter (Biomonitor S.C.I. AB, UmeÁ¥, Sweden) with light level 200 lmol photons m-2 second-1. The indexes measured were the ratio of variable to maximum fluorescence yielded under the artificial light treatment (Fv/Fm) and the time needed for the leaf to reach half of its Fm (T1/2). To make sure that maximum level of fluorescence was reached, the leaves were dark adapted for 15 minutes with a lightweight leaf cuvette. Larger values of Fv/Fm and smaller values of T1/2 were considered to be representative of faster growth and less stress and thus higher fitness (Á–quist & Wass 1988).

Ecological Outcomes Achieved

Eliminate existing threats to the ecosystem:
Both boxing and full shelter reduced wind velocity to 40 and 73% of control, respectively. Similarly, boxing and full shelter both reduced illumination to 56 and 92% relative to control, respectively. None of the treatments influenced the amount of rainfall reaching the ground. Liming had a strong positive effect on birch seedlings; only seedlings in liming treatments survived until 2004 at site 1 (mortality 60%), and at site 2, average survival in liming treatments was double, relative to control by 2004. At site 1, all control seedlings were dead by 2001, and when analyzing survival until 2000, only liming had a significant positive effect. Moreover, the growth of survivors was enhanced, as indicated by larger leaf size, increased long shoot length (by 81%), and tripled vertical increment. Mitigation of environmental stress by liming was also indicated by lower T1/2 values, although Fv/Fm values were not affected. Effects of boxing were generally non-significant, although in the first year, boxing increased leaf length by 15%. The only consistent positive effects were detected on Fv/Fm and T1/2. Fv/Fm increased by 30% in 2000 and 13% in 2001, and T1/2 decreased by 24% in 2000 and by 12% in 2001. Marginally significant interactions were found between liming and boxing on seedling height and T1/2, with the two treatments being less beneficial together than separately. Site had a pronounced effect on most of the investigated variables; both survival and growth were lower at site 1. No differences were found between summer and winter mortality. The effects of both treatments were differently expressed at the two study sites. In seedling height analysis, year x site x liming interaction was marginally significant, with liming benefiting seedlings only at site 2, increasing height by 58, 176, and 243% in 2000, 2001, and 2004, respectively. Although partial shelter (PS) and stone (St) increased seedling survival at site 2, only full shelter (FS) had a statistically significant beneficial effect on seedling performance. Full shelter increased leaf length by 11% and height by 12% relative to control. Also, both partial shelter and stone had a marginally significant positive effect on leaf length, both increasing it by 10% relative to control. All dependent variables showed pronounced inter-year variation at site 2. At site 1, the inter-year pattern was generally the same as at site 2; however, the differences were not significant, presumably due to the low number of survivors. Neither performance nor survival until 2000 was increased by any mode of sheltering at site 1.

Factors limiting recovery of the ecosystem:
In spite of greatly reducing wind stress experienced by the seedlings, boxing had no positive effect on survival and had only a small positive effect on performance during the first growing season. Only the chlorophyll fluorescence measurements showed a decrease in stress in the boxing treatment. These results are in contrast with the conclusions of an earlier study conducted in the same site by Kozlov & Haukioja (1999), which demonstrated a positive effect of boxing on the performance and survival of Mountain birch seedlings. The most plausible explanation for the discrepancy in the results of the two experiments is that the emissions from Severonikel smelter have decreased since the time of the previous study (Barcan 2002). Thus, the effect of airborne pollution, acting synergistically with wind stress (Hoad et al. 1998), relative to soil pollution has decreased. Also, among-year variation in weather conditions may have contributed to the differences in results. Finally, we monitored the birch seedlings for five growth seasons compared to only two growth seasons in the earlier study (Kozlov & Haukioja 1999); boxing indeed increased leaf length in the first year of the experiment, but this beneficial effect was gone the following years. There are also other possible explanations for the low effect of boxing compared to the other shelters: (1) in early winter the drifting snow may accumulate outside, not inside the boxes, and thus the seedlings can be vulnerable to freezing and (2) microclimate changes caused by boxing (e.g., shading) may have had negative effects on the birch seedlings. Shading is expected to increase plant growth, and this has been shown to be the case also with birch seedlings (Aphalo & Lehto 1997; Van Hees & Clerkx 2003). However, an increase in growth may have accelerated heavy metal accumulation and reduced survival. If the fastest growing individuals were the first to die, then this could explain why the beneficial effects of boxing were not evident in height data either. However, boxing did not increase growth or survival even in conjunction with liming, which casts doubt on the last explanation. It must be noted, however, that the limed seedlings outgrew the boxes (25 cm high) by the end of the experiment. During the last years of the experiment, both the protection offered and the shading imposed by boxing may have been negligible. The chlorophyll fluorescence data give some support to this explanation, with the beneficial effects of boxing being twice as strong in 2000 as in 2001. It is, however, important to note that shading is known to increase the proportional importance of photosystem II compared to photosystem I (BjÁ¶rkman 1981; Anderson & Osmond 1987). Because chlorophyll fluorescence measures plant condition via the efficiency of photosystem II, it may simply be that the chlorophyll fluorescence measurements reflected the effects of shading on the physiology of Mountain birch seedlings, even though other factors were more critical to their survival and performance.

Socio-Economic & Community Outcomes Achieved

Key Lessons Learned

The data obtained in this study provided partial support for the hypothesis that mitigation of physical stressors by sheltering increases the performance Mountain birch seedlings planted in a heavily polluted barren site that has soil nickel and copper concentrations of up to 4,000 and 2,000 μg/g, respectively (Barkan 1993). Although liming produced stronger positive effects, the beneficial effects of full shelter, partial shelter, and stone on survival and performance were evident at site 2. The positive effects of a natural shelter (stone) were of similar magnitude as with artificial shelters. At site 1, where winter snow cover was thinner and soil water content lower than in site 2, sheltering was not sufficient to overcome these stressors and seedling mortality was 100%.

Absence of differences between summer and winter mortality suggests that the positive effects of sheltering are not restricted to the growing season. Various shelters are likely not only to mitigate wind stress during the growth season but also to protect seedlings from freezing via trapping snow drifting over the barren site at the beginning of winter. Causality is further obscured by the fact that protection from freezing not only improves winter survival but can also increase seedling performance during the following summer (Weih & Karlsson 2002) and vice versa: good summer conditions positively affect winter survival (Weih & Karlsson 1999).

As expected, liming had a strong positive effect on Mountain birch survival and performance, verifying the negative effects of soil contamination on the seedlings. Even the small doses used in the current experiment, 40 g/m2 compared to 1,000 g/m2 in the Sudbury restoration program (Lautenbach et al. 1995), increased the growth of birch seedlings 3-fold. Heavy metal toxicity is often associated with a deficiency of essential nutrients and problems in water uptake (Jones & Hutchinson 1988; Barcelo & Poschenrieder 1990; Kozlov et al. 1999). Thus, the effects of liming were most likely not restricted to reduction of soil toxicity alone. This view is supported by the fact that liming was especially crucial at site 1 and sheltering was beneficial only at site 2, although site 1 is less polluted but dried and more oligotrophic than site 2.

In conclusion, the results of the study indicate that under the environmental conditions and current level of emissions (circa 40 kt of SO2 annually) at the experimental site, soil quality/toxicity influences the long-term performance of Mountain birch more than physical stress. Consequently, liming, which reduces soil toxicity and increases its nutritional quality, is clearly needed to assure successful restoration. This does not mean, however, that positive effects could not be achieved by physical sheltering. Indeed, sheltering, even by natural windbreaks such as large stones, may well be a useful tool in reforestation when the environment is not so harsh as to cause 100% mortality during the first few years of plant development and when liming is for some reason not a feasible option.

Long-Term Management

The data gathered in this study imply that when assessing plant performance in polluted habitats, it is important to monitor the plants for several years. Survival data from site 2 illustrate this statement exceptionally clearly: no treatments had any effect on survival by the year 2000, and only by 2004 did the effects of various treatments become apparent. Another finding is the need to measure several performance parameters because they can respond to treatments differently. Chlorophyll fluorescence–which is a widely used index in the assessment of plant stress, including stress imposed by pollution (Adams et al. 1989; Kitao et al. 1997; Odasz-Albrigtsen et al. 2000), and has been shown to correlate strongly with the CO2 metabolism of a wide range of plants, including birch (Krause & Weis 1991; Ball et al. 1995; Govindjee 1995)–gave very different results from growth and survival measurements. In this experiment, the seemingly low stress in boxing treatments (as indicated by chlorophyll fluorescence data) was not evident in growth parameters or survival. It may be that the beneficial effects of the boxing treatment (e.g., alleviated wind stress or protection from freezing) were overcome by excess growth, accumulation of pollutants, or some other factors. Another possibility is that shading had no effect on performance characteristics or survival but merely affected chlorophyll fluorescence measurements via the ratio of photosystems I and II as discussed earlier. In any case, had the analysis of results relied solely on chlorophyll fluorescence data or growth data, the interpretations of this study would have been drastically different.

Sources and Amounts of Funding

Funding was received from the Maj and Tor Nessling Foundation (1999 – 2001) and the Academy of Finland (research project 201991).

Other Resources

ErÁ¤nen, Janne K. and Mikhail V. Kozlov. 2006. Physical sheltering and liming improve survival and performance of mountain birch seedlings: a 5-year study in a heavily polluted industrial barren. Restoration Ecology 14(1): 77-86.

Kozlov, Mikhail V. And Valery Barcan. 2000. Environmental contamination in the central part of the Kola Peninsula: history, documentation, and perception. Ambio 29(8).

Primary Contact

Organizational Contact